[extropy-chat] sjbrain calcs

[Spike]

YEEESSSS!  Robert you are THERE, bud!

> On Sun, 21 Dec 2003, Spike wrote:
> 
> > If we had an SJBrain, the space in
> > which the Jupiter mass of nodes resides is much smaller
> > than the space occupied by a true Jupiter-MBrain, half a
> > cubic light minute
> 
> Either I still don't understand this picture Spike or there is
> something wrong with your calculations or Anders.
> Anders' Jupiter Brain (Zeus) has a radius of ~9000 km (~1.5
> Earth radii) and mass of ~10^25 kg (~1.8 Earth masses).

Yes, but this is a Jupiter Brain, not the MBrain.  In Anders'
JBrain, the whole device completely solid, given the numbers you
quote above.  The brilliance of the MBrain is that it does not
have hot material anywhere nor material under enormous pressure 
at all.  It is a stunningly enormous number of individual particles, 
cheerfully orbiting and calculating away in parallel with the 
other nodes.

> Its probably going to require much of the carbon (for diamondoid)
> that is in the inner planets...

Again I agree, and further speculate that any extremely large
node (more than a few kg) would result in material being
wasted, material not available for computation.  If the node
is that large, heat control becomes an issue.  If the node
is a microgram scale, there is plenty of surface area from 
which to radiate heat generated in computation.

> 
> > as opposed to Robert's 250,000 cubic light minutes.
> > The notion of collision control in an SJBrain
> > is more difficult.
> 
> You must be pulling these numbers out of the air spike.  I don't
> think I ever specified the size of an MBrain...

We did, Robert, about 2 years ago when you were here.  I 
still have the green notebook in which we did the calcs.  
OK I have it open before me now.  Tuesday 29 January 2002.  
Were you not in Silicon Valley Taxifornia on the date in 
question, Mr. Bradbury?  {8^D

We theorized a .1 gram node, calculated some orbit mechanics
assuming concentric rings of nodes one meter apart and the
rings nested on one meter intervals (each ring a meter larger
radius than the one immediately inside it) and each ring
tilted a microradian from the next adjacent ring, so no
intersecting orbits anywhere.

This causes me to realize why Gene might be choking on the
concept.  The term rings might cause mental pictures of
Saturn's rings, which are flattened disc thingies.  The rings 
we theorized are thin strings of nodes spaced at one meter 
apart and circling the sun in an enormous precession.  For
instance the ring that is 150,000,000,000 meters from the sun,
(about the earth's orbit distance) would contain 942,477,796,077
nodes.  The ring a meter sunward, 942,477,796,071 nodes and
the ring a meter anti-sunward 942,477,796,083 nodes.  See how
I got those numbers?

>  If my calc's are
> right 250,000 culm gives you a radius of 39 light-minutes 
> which is around the asteroid belt someplace...

Ja, and the reason is because yesterday I was estimating
an MBrain with outer limits around Jupiter, and I recalled 
that Jupiter orbits in about 12 years, so 12^(2/3) is about
5 AU, and an AU is about 8 minutes, so Jupiter is
about 40 minutes out there, so a sphere that size would
occupy about 250,000 cubic minutes.  You can practically
do this stuff in your head, Robert.

An SBrain uses the same amount of material but is packed
much tighter.  Not nearly as tight as an Sandbergian JBrain,
but tighter than a traditional MBrain.  The Jupiter mass
of nodes orbit 5 AU from the central star like an enormous
fluffy planet out there by itself, over 100 times the diameter
of Jupiter.  Think of it as analogous to the guy at the carnival
converting a sugar cube a cm on a side into a big fluffy wad 
of cotton candy.

Do tell me you guys are old enough to have seen cotton
candy some time in your lives. 

>  Actually if you can avoid the people
> on Earth screaming you start building an MBrain way inside the
> orbit of Mercury using TiC as the primary material...

Ja but of course my scenario is post-uploading, so the
only screaming you would hear is the wild cheering of
those already occupying the existing computronium, who
are witnessing the transformation of dull, dead raw material
into vibrant thinking living useful creative matter.

Robert, your MBrain is the greatest idea of the 1990s.  It's
humanity's future home bud, the traditional heavenly kingdom,
except without the deity and the harps.  If one concludes as
I have, that the singularity is inevitable, then the MBrain 
is the logical next step: we would need to grab up all the
available metals in the solar system and get it all THINKING.
Its the ultimate order from chaos.  

I cannot even imagine a long-term future without some form 
of singularity, and *every* singularity scenario I can imagine 
eventually leads to an MBrain, or perhaps the more specific 
form of the MBrain, the SBrain.  So *every* post-singularity
star would have all its available metal in the form of an
MBrain.  So the original question is a profound one indeed: 
can the SIRTF detect an MBrain?

spike